numpy实现两层layer

暑期第二天回顾两层网络训练细节，对应CS231n-class-10，练习numpy写layer

 1 import numpy as np
 2 
 3 x = np.array([[0, 0, 1], [0, 1, 1], [1, 0, 1], [1, 1, 1]])
 4 y = np.array([[0, 1, 1, 0]]).T
 5 print(x)
 6 print(y)
 7 
 8 w0 = 2*np.random.random((3, 4)) - 1
 9 w1 = 2*np.random.random((4, 1)) - 1
10 
11 for j in range(60000):
12     l1 = 1 / (1 + np.exp(-(np.dot(x, w0))))
13     l2 = 1 / (1 + np.exp(-(np.dot(l1, w1))))
14     l2_delta = (y-l2)*(l2*(1-l2))
15     l1_delta = l2_delta.dot(w1.T) * (l1*(1-l1))
16     w1 += l1.T.dot(l2_delta)
17     w0 += x.T.dot(l1_delta)
18     loss=y-l2
19     if(j%10000==0):
20         print(loss)
21 print(l2)

 

转载于:https://www.cnblogs.com/wind-chaser/p/11158847.html